Accuracy and Privacy Evaluation of detected communities using Attributed-Based Label Propagation Method

Community Detection plays an integral part in network analysis, as it facilitates understanding the structures and functional characteristics of the network. Attribute-Based Label Propagation method ABLP was proposed by the authors which is a Modularity maximization based on Label Propagation algorithm with regards to homogeneity. The main concept is in each run, the Modularity measure is calculated, and the maximum value is considered, and at the same time, examining the nodes' attributes to assign similar nodes in a single community. In this paper, the ABLP method is evaluated in terms of accuracy and privacy measures such as (NMI, ARI, RI, VI, and Split- Join distance).